Since the early 1960s

several countries where earthquake occurred frequently have carried out continuous researches into earthquake prediction in a planned way. However

on the whole

there still exists a great distance between the advance achieved and the scientific goal of realizing earthquake prediction. This reminds people of difficulties in earthquake prediction

which is much greater than originally expected. In view of the various difficulties to which earthquake prediction has often to be faced

scientists around the world generally accept the fact that earthquake prediction needs long-term scientific explorations and scientific accumulations. On the other hand

however

earthquake prediction is facing to the increasingly strong social demand. According to relevant data we can find that the more developed the social economy

the more serious the disaster caused by earthquakes. The work of precautions against earthquakes and disaster alleviation has become an important link between science and technology and economic social development. In the late 1990s

some large international research projects were progressively implemented. The outcomes obtained from these projects contribute a new and important basis for the development of earthquake prediction. Through reviewing the work of the past 40 years and focusing on the problems of sustainable development in this field

this article summarizes several questions about earthquake prediction

which deserve attention from now on. （a） Combination of the practice of earthquake prediction with the researches on basic theory should be very important

（b） It is needed to emphasize the primary role of seismic observations in this research field

（c） We should keep constructing test sites for earthquake prediction and give full play to their positive role

（d） Among long-term

medium-term

short-term and impending earthquake predictions

the former should acts as a leading method

（e） Scientific groups should be built up and trained for the purpose of earthquake prediction.